Which is better? PCM or Microsoft ADPCM ?
Well, that all depends on your processor speed and memory, and what kind of audio quality you deem important, among other things! Here is the “whole truth” concerning both!
PCM (which stands for “Pulse Code Modulation”) is GENERALLY SPEAKING the same as 16-bit “CD” audio (as most people use it, though it is available in 8-bit format as well – see below), and it is the standard, having no compression at all – every snapshot, from silence to the loudest waveform burst, is represented by a full 16-bit “word”. It will exhibit the highest audio-portion file size. It records audio into the .wav format into a linear stream of samples.
“Because these must be stored as finite-precision digital numbers, the data is truncated to either 16-bit PCM or 8-bit PCM, commonly called 8- and 16-bit samples. 16-bit data has more resolution, so the digital waveform sounds better. 8-bit PCM has less resolution, causing audible hiss in the waveform. It also requires less disk space. ” – from Microsoft: http://support.microsoft.com/kb/89879
Microsoft ADPCM (which stands for “Adaptive Delta Pulse Code Modulation” – a mouthful!) allows other bit depths and sample rates to be selected, and uses only 4 bits per sample (audio snapshot), but can move these 4 bits around to represent differing volume levels (the delta part moves the “sector” being represented by the 4 bits up and down as a range of volume levels being represented). The fewer the number of bits, the less information. The lower the sample rate, the fewer “snapshots” per second are taken of the audio, and so the less information again. You pick what combination sounds acceptable. However,
“Because the Windows Sound System hardware only understands 8/16-bit PCM, the computer must compress and decompress the ADPCM into/from PCM, which requires CPU time.” – from Microsoft: http://support.microsoft.com/kb/89879 So, if being done “on the fly” as you record, it will slow down your computer. That said, a more powerful CPU may be able to handle this load without causing any lagging of the audio vs. video at all. (Also from the Microsoft site: “ADPCM stores the value differences between two adjacent PCM samples and makes some assumptions that allow data reduction. Because of these assumptions, low frequencies are properly reproduced, but any high frequencies tend to get distorted. The distortion is easily audible in 11 kHz ADPCM files, but becomes more difficult to discern with higher sampling rates, and is virtually impossible to recognize with 44 kHz ADPCM files.”)
If using CamStudio to record videos, I generally always use the “Use MCI for Recording” checkbox being selected (which records using the system’s native PCM method); I live with the large, 16-bit audio file produced, and use VirtualDub ( http://virtualdub.sourceforge.net/ ) to compress the audio after making my video before I upload it to YouTube. (Have Video set to “Direct Copy Mode” and Audio set to “Full Processing”, then pick your compression – I use LAME, which is a free download: http://www.free-codecs.com/download/lame_acm_codec.htm ).